Proc. Natl. Acad. Sci. USAVol. 76, No. 1, pp. 333-337, January 1979Cell Biology

Binding site on macrophages that mediates uptake and degradationof acetylated low density lipoprotein, producing massivecholesterol deposition

(cell surface receptors/denatured proteins/cultured cells/lysosomes/familial hypercholesterolemia)

JOSEPH L. GOLDSTEIN, Y. K. Ho, SANDIP K. BASU, AND MICHAEL S. BROWNDepartments of Molecular Genetics and Internal Medicine, University of Texas Health Science Center at Dallas, 5323 Harry Hines Blvd., Dallas, Texas 75235

Communicated by George E. Palade, October 16, 1978

ABSTRACI Resident mouse peritoneal macro hages wereshown to take up and degrade acetylated 1251 abeled lowdensity lipoprotein (15I-acetyl-LDL) in vitro at rates that were20fold greater than those for the uptake and degradation of125I-LDL. The uptake of 1251-acetyl-LDL and its subsequentdegradation in lysosomes were attributable to a high-affinity,trypsin-sensitive, surface binding site that recognized acetyl-LDL but not native LDL. When 25I-acetyl-LDL was bound tothis site at 4VC and the macrophages were subsequently warmedto 370C, 75% of the cell-bound radioactivity was degraded tomono[125I]iodotyrosine within 1 hr. The macrophage bindingsite also recognized maleylated LDL, maleylated albumin, andtwo sulfated polysaccharides (fucoidin and dextran sulfate) in-dicating that negative charges were important in the bindingreaction. A similar binding site was present on rat peritonealmacrophages, guinea pig Kupffer cells, and cultured humanmonocytes but not on human lymphocytes or fibroblasts, mouseL cells or Y-1 adrenal cells, or Chinese hamster ovary cells. Up-take and degradation of acetyl-LDL via this binding site stim-ulated cholesterol esterification 100-fold and produced a 38-foldincrease in the cellular content of cholesterol in mouse perito-neal macrophages. Although the physiologic significance, if any,of this macrophage uptake mechanism is not yet known, wehypothesize that it may mediate the degradation of denaturedLDL in the body and thus serve as a "backup" mechanism forthe previously described receptor-mediated degradation ofnative LDL that occurs in parenchymal cells. Such a scavengerpathway might account for the widespread deposition ofLDL-derived cholesteryl esters in macrophages of patients withfamilial hypercholesterolemia in whom the parenchymal cellpathway for LDL degradation is blocked, owing to a geneticdeficiency of receptors for native LDL.

Studies in cultured human fibroblasts and blood lymphocyteshave defined a receptor-mediated pathway by which cells takeup and degrade low density lipoprotein (LDL), the majorcholesterol-transport protein in human plasma (1, 2). This LDLuptake pathway is believed to function primarily in extrahe-patic parenchymal cells where it serves a dual function: itprovides the major physiologic route of LDL degradation inthe human body and, at the same time, it supplies cholesterolto cells for the synthesis of membranes and steroid hor-mones.

Patients with familial hypercholesterolemia (FH) have amutation in the gene specifying the LDL receptor (1, 2). Inthese patients, LDL degradation through the physiologicpathway is blocked, and each LDL particle survives in thecirculation 2-3 times longer than it does in normal subjects(3-5). As a result, the lipoprotein accumulates to abnormallyhigh levels in plasma. Eventually the LDL is degraded throughan alternate pathway. Inasmuch as the excess LDL-cholesterolbecomes deposited in nonparenchymal cells throughout the

body, including hepatic Kupffer cells and macrophages of thespleen, kidney, bone marrow, skin, tendons, and other organs(2, 6, 7), we have postulated that the alternate pathway for LDLdegradation is expressed primarily in such scavenger cells (2,5). Because this scavenger mechanism acts on plasma LDL thathas circulated for an abnormally long time in vio, it mayrecognize some form of LDL that has become denatured orchemically modified. The clearance of denatured proteins invivo by macrophages is well known (8, 9).

As a first step in a search for the postulated macrophagesystem mediating the uptake and degradation of modified ordenatured LDL, we have used as a probe 125I-labeled LDL(125I-LDL) that has been extensively acetylated in vitro(125I-acetyl-LDL). The rationale for this choice was based onprevious studies showing that acetylation of LDL altered thelipoprotein to the extent that it was no longer recognized by theLDL receptor in fibroblasts (10). As a model cell system forstudying the metabolism of '25I-acetyl-LDL, we have usedresident mouse peritoneal macrophages isolated by the tech-niques developed by Cohn and coworkers (11, 12). We dem-onstrate that these cells possess a high-affinity binding site thatrecognizes acetyl-LDL but not native LDL. This binding site,which so far appears to occur only on macrophages, mediatesthe rapid uptake and degradation of acetyl-LDL, producingmassive cholesterol deposition in these cells.

METHODSMouse Peritoneal Macrophages. Female NCS mice (25-30

g) were obtained from The Rockefeller University. Peritonealcells were harvested from unstimulated mice in Dulbecco'sphosphate-buffered saline (cat. no. 310-4190, GIBCO) (11, 12).The fluid from 12-20 mice (6-10 X 106 cells per mouse) waspooled, and the cells were collected by centrifugation (400 Xg, 10 min, 240) and washed once with 30 ml of Dulbecco'smodified Eagle's medium (henceforth referred to as medium).The cells were resuspended in medium A (medium containing20% fetal calf serum and 100 units of penicillin and 100 ,ug ofstreptomycin per ml) at 2 X 106 cells per ml. Aliquots (1 ml)were dispensed into 35 X 10-mm plastic petri dishes and in-cubated in a humidified CO2 (5%) incubator at 37°C. After 2hr each dish was washed three times with 2 ml of mediumwithout serum. Unless otherwise stated, the medium was im-mediately replaced with 1 ml of medium B (medium con-taining 10% human lipoprotein-deficient serum), and the ex-periment was initiated. Each dish of adherent macrophagescontained about 30% of the total number of cells originallyplated (30-40 jig of cell protein).

Other Cells. Human fibroblasts, mouse L cells, mouse Y-1adrenal cells, and Chinese hamster ovary cells (CHO-Ki) were

Abbreviations: FH, familial hypercholesterolemia; LDL, low densitylipoprotein.

333

The publication costs of this article were defrayed in part by pagecharge payment. This article must therefore be hereby marked "ad-vertisement" in accordance with 18 U. S. C. §1734 solely to indicatethis fact.

334 Cell Biology: Goldstein et al.

grown in monolayer in either Eagle's minimum essential me-dium (13) or Ham's F-12 medium containing 10% fetal calfserum. Human lymphoblasts (1) were grown in suspension inRPMI 1640 medium containing 10% fetal calf serum. Ratperitoneal macrophages were isolated as described for mousemacrophages. Monolayers of guinea pig Kupffer cells were

prepared by Thomas Rogoff by a modification of the methodof Garvey (14). Human mononuclear cells were isolated fromperipheral blood of a healthy subject by the method of Boyum(15), and the monocytes and lymphocytes were separated bydifferential adherence to plastic (16). The monocytes (adherentcells) and lymphocytes (nonadherent cells) were incubatedseparately for 5 days in RPMI 1640 medium containing 10%autologous human serum (16).

Lipoproteins. LDL (p, 1.019-1.063 g/ml) and lipoprotein-deficient serum (p > 1.215 g/ml) were isolated from plasmaof individual subjects by differential ultracentrifugation (10).The concentration of LDL is given in terms of its protein con-

tent.Chemical Modification of Proteins. LDL and human serum

albumin (Cohn fraction V) were acetylated with repeated ad-ditions of acetic anhydride (17) as described (10). Acetic an-

hydride was added at 40-fold molar excess with regard to totallysines in LDL. LDL and albumin were maleylated by repeatedadditions of maleic anhydride (18). All of the acetylated andmaleylated proteins had enhanced mobility on electrophoresisin agarose gel at pH 8.6 (see figure 1 in ref. 10). The particle sizeof acetyl-LDL was similar to that of LDL by electron micros-copy after negative staining with uranyl acetate. LDL, ace-

tyl-LDL, and maleyl-albumin were labeled with 1251 as de-scribed (10).

Assays. The total cellular content of 125I-LDL and 125I-acetyl-LDL in mouse macrophage monolayers was measuredat either 4 or 370C as described for 125I-LDL in fibroblasts (13)except for three modifications: (i) the volume of each of the sixwashes was 2 ml rather than 3 ml; (ii) the heparin-release stepwas omitted; and (Mi) after the standard wash, the cells weredissolved by incubation at 240C for at least 1 hr in 0.6 ml of 0.2M NaOH. The whole sample was assayed to determine the 125Iradioactivity associated with the cells, after which an aliquotwas used to determine the content of cellular protein (19). Theproteolytic degradation of 125I-LDL, 125I-acetyl-LDL, and125I-labeled maleyl-albumin was measured by assaying theamount of 125I-labeled trichloroacetic acid-soluble (noniodide)material formed by the cells and excreted into the culturemedium (20). The incorporation of [14C]oleate into cellularcholesteryl ['4C]oleate (21) and the cellular content of free andesterified cholesterol (22) were measured by the referencedmethods.

RESULTSWhen freshly isolated mouse peritoneal macrophages were

incubated with increasing concentrations of 125I-acetyl-LDLat 37°C, the cellular content of radioactivity increased in a

saturable fashion (Fig. IA). During the 5-hr incubation, a

portion of the added 125I-acetyl-LDL was degraded to tri-chloroacetic acid-soluble material (Fig. 1B). This material wasshown by paper chromatography to consist entirely ofmono[125I]iodotyrosine. As the 125I-acetyl-LDL concentrationincreased, the rate of degradation increased hyperbolically inparallel with the saturable component of the cellular uptakeprocess; in both cases, half-maximal values were achieved atan 125I-acetyl-LDL concentration of about 25 ,ug/ml. In con-

trast, only small amounts of 125I-LDL became associated withthe macrophages (Fig. 1A). This uptake showed no evidenceof saturation. Moreover, only small amounts of 125I-LDL were

0°- 1.2-

080,':- E0.8-

_Cm

L-m 0.4-

.5

0

cn,4b0,0

C.

a'D

0. tI-

V)

0 50 100 150 0 50 100 150'25 -Lipoprotein in medium, gg protein/ml

FIG. 1. Accumulation and degradation of 125I-LDL (A) and1251-acetyl-LDL (@) by mouse peritoneal macrophages. Each dishreceived 1 ml of medium B containing the indicated concentrationof either 125I-LDL (57,000 cpm/tg of protein) or 1251-acetyl-LDL(50,000 cpm/gg of protein). After incubation for 5 hr at 37°C, theamount of 125I-lipoprotein in the cells (A) and the amount of 1251_labeled acid-soluble material in the medium (B) were determined induplicate dishes.

degraded by the freshly isolated mouse macrophages (Fig.1B).

In macrophages incubated with 125I-acetyl-LDL at 37°C forvarying lengths of time, the cellular content of radioactivityreached a steady-state plateau after 1 hr (Fig. 2A). After thissteady-state was reached, acid-soluble radioactivity continuedto appear in the medium at a linear rate, reflecting the con-

tinuing uptake and degradation of the '25I-acetyl-LDL. After5 hr, approximately 9 times as much '25I-acetyl-LDL had beendegraded as was contained within the cells during the steady-state. In the presence of the lysosomal inhibitor chloroquine (1),degradation was abolished, and the cellular content of 125I-acetyl-LDL increased linearly for 5 hr (Fig. 2B).To measure the surface binding of 125I-acetyl-LDL, we in-

cubated the macrophages with increasing concentrations of125I-acetyl-LDL at 40C. The 125I-acetyl-LDL bound to the cellswith saturation kinetics (Fig. 3A). Maximal binding was

achieved within 120 min at both 5 and 50 ,ug/ml. Two-thirdsof the 125I-acetyl LDL bound at 4°C could be released bysubsequent incubation of the cells with trypsin (1 mg/ml) at40C. As with native LDL in fibroblasts (13), the half-maximalconcentration for 125I-acetyl-LDL binding in macrophages at

EI

cm

.3

J-i

a

Uj

A B

10'A

8

6-

A4 A-2 A/

0 12 5 012 3 4O 1 2 3 4 5 O 1 2 3 4 5

Time, hr

FIG. 2. Effect of chloroquine on time course of accumulation (@)and degradation (A) of 1251-acetyl-LDL by mouse peritoneal macro-phages. The cells were prepared by the standard method except thatthe monolayers were incubated for 18 hr in 1 ml of medium A priorto the experiment. Each monolayer was then washed with 2 ml ofmedium, after which 1 ml of medium B containing 1251-acetyl-LDL(10 ,g/ml; 98,000 cpm/,ug of protein) without (A) or with (B) 75 ;Mchloroquine was added. After incubation at 37°C for the indicatedinterval, the amount of 1251-labeled acid-soluble material in the me-dium and the amount of 1251-acetyl-LDL in the cells were determinedin duplicate dishes.

Proc. Natl. Acad. Sci. USA 76 (1979)

Proc. Natl. Acad. Sci. USA 76 (1979) 335

*0

C50-J

-0

0Cu

Cell Biology: Goldstein et al.

I B

W~~fly~1

cm

160

120 _30-j

80 -

40In

U0 25 50 75 1000 30 60 90 120 180

'25 -Acetyl-LDL, Time at 370C, minMug protein/ml

FIG. 3. (A) Binding of 1251-acetyl-LDL by mouse peritonealmacrophages at 40C. Each dish received 1 ml of ice-cold medium Band was placed at 40C for 30 min, after which the indicated concen-tration of '251-acetyl-LDL (734 cpm/ng protein) was added. Themonolayers were incubated at 40C for 2 hr and then washed, and theamount of 1251-acetyl-LDL bound to the cells was determined in du-plicate dishes. (B) Degradation at 370C of 125I-LDL previously boundby mouse peritoneal macrophages at 40C. Each dish was incubatedfor 2 hr at 40C with 1251-acetyl-LDL (10 ptg/ml; 738 cpm/ng protein)as in A, after which each monolayer was washed by the standardprocedure. Each dish then received 1 ml of warm medium B con-

taining 20 ,4g of unlabeled acetyl-LDL per ml, and all the dishes wereincubated at 370C. At the indicated interval, duplicate dishes were

rapidly chilled to 41C, the medium was removed, and its content of1251-labeled acid-soluble material (A) was measured. The total amountofL251-acetyl-LDL that remained associated with the cell (0) was alsodetermined.

40C (5 ttg/ml) was lower than the half-maximal concentrationfor cellular uptake at 370C (25 ,ug/ml, Fig. 1A). In anotherexperiment, cells that had bound 125I-acetyl-LDL at 40C were

subsequently warmed to 370C (Fig. 3B). The cell-bound ra-

dioactivity declined rapidly. Seventy-five percent of the declinewas due to the excretion of acid-soluble radioactivity from thecell, a reaction that reached completion by 60 min. The 25%of cell-bound 125I-acetyl-LDL that was not degraded dissociatedfrom the cell surface soon after warming and appeared in themedium as acid-precipitable material. Such dissociation didnot occur at 4VC.

Fig. 4 shows experiments in which macrophages were in-cubated with 125I-acetyl-LDL in the presence of increasingconcentrations of potential competitors for the binding site.LDL failed to compete with the 125I-acetyl-LDL for degra-dation. On the other hand, unlabeled acetyl-LDL was an ef-fective competitor as was maleyl-LDL (Fig. 4A). Acetyl-al-bumin failed to compete with 125I-acetyl-LDL; however,maleyl-albumin was an effective competitor (Fig. 4B).

Inasmuch as acetyl-LDL, maleyl-LDL, and maleyl-albumin

e0 125- A B. *C

_Ac00

o 50 MleylLD albu*Miny-[ cd

(D-glutamic

_jc 725 k ,9\\acid)

o 50 bymoMaleyl-LDL Maleyl-

25m/ngopeiAcetyl- albumin At Fucoidin25~~ ~ ~ ~ ~ ~ \/Dextran

5 hr at37CHhmuto 2Ilaee cdslbemtralnthe

M 0 !7*0 150 300 450 0 150 300 450 0 25 50 500 2000

Competing compound in medium, pg/ml

FIG. 4. Ability of various compounds to inhibit the degradationof "251-acetyl-LDL by mouse peritoneal macrophages. Each dish re-

ceived 1 ml of medium B containing 1251-acetyl-LDL (25 jig/ml; 60-70

cpm/ng of protein) and the indicated compound. After incubation for

5 hr at 370C, the amount of 12,51-labeled acid-soluble material in the

medium was determined. The 100% values for degradation of 1251_acetyl-LDL in the absence of competing compounds were 4600,4355,and 11,500 ng- 5 hr-'-mg of protein-' for A, B, and C, respectively.

have net negative charges, the data suggested that negativecharges play a role in the binding reaction. This conclusion wassupported by the finding that dextran sulfate (Mr 500,000) andfucoidin, two sulfated polysaccharides, both competed effec-tively with 1251-acetyl-LDL for degradation. However,poly(D-glutamic acid) (Mr 50,000-100,000), which also has anegative charge, failed to compete (Fig. 4C). In each of theexperiments of Fig. 4, the amount of cell-bound 125I-acetyl-LDL declined in a manner parallel to that of the degradationrate. Moreover, each of the compounds in Fig. 4 was tested asan inhibitor of 1251-acetyl-LDL binding to the cells at 40C, andagain the results paralleled those shown in Fig. 4. That theseinhibitors were acting by binding to the cells and not by bindingto 125I-acetyl-LDL was indicated by the finding that prior in-cubation of cells with fucoidin or dextran sulfate at 40C fol-lowed by washing the cells inhibited the subsequent 125I-acetyl-LDL binding by more than 90%.

Neither dextran (Mr 500,000) nor three sulfated polysac-charides (heparin and chondroitin sulfates A and C) inhibited125I-acetyl-LDL binding or degradation when tested at con-centrations up to 500.,gg/ml. Other negatively charged com-pounds, such as RNA, phosvitin, colominic acid (polysialic acid),and polyphosphates (chain length, 65), were also ineffectiveinhibitors at 500 Aig/ml. Other ineffective inhibitors included:lysozyme, thyroglobulin, fetuin, orosomucoid, and asialo-oro-somucoid at concentrations of 500-5000,gg/ml; whole humanserum at 30 mg of protein per ml (50% serum); and mannan(from bakers yeast) at 5 mg/ml.

1251-Maleyl-albumin (25 Atg/ml) was degraded by the mouseperitoneal macrophages at a rate (4.7 jtg-5 hr'.-mg protein-')similar to that for 1251-acetyl-LDL. The degradation of 1251-maleyl albumin was inhibited by more than 80% by a 20-foldexcess of unlabeled inaleyl-albumin and acetyl-LDL, con-firming that these two substances were taken up through thesame binding site. A 20-fold excess of native albumin did notcompete for the degradation of 12,I-maleyl albumin or121acetyl-LDL.The amount of high-affinity binding of 125I-acetyl-LDL at

40C was reduced by 90% when the macrophages were firstincubated for 50 min at 370C with either trypsin (250 jzg/ml)or Pronase(10Eug/ml).

High-affinity degradation of 125I-acetyl-LDL occurred notonly in mouse peritoneal macrophages but also in rat peritonealmacrophages, guinea pig Kupffer cells, and cultured humanblood monocytes (Table 1). These phagocytic cells degraded12,5-acetyl-LDL 3.8-177 times more rapidly than they de-graded 1251-LDL. On the other hand, none of the nonmacro-phage cells showed significant high-affinity degradation of1251-acetyl-LDL. In fact, all the nonmacrophage cells testedshowed much greater rates of degradation of 1251-LDL thanof 1251-acetyl LDL. When the nonmacrophage cells were in-cubated in the absence of lipoproteins for 48 hr, a conditionknown to increase LDL receptor activity (1), the rate of deg-radation of 1251-LDL increased 5- to 15-fold above the valuesshown in Table I, whereas the degradation of 125I-acetyl-LDLdid not change.

In fibroblasts the degradation of LDL releases free choles-terol, and this in turn enhances the rate at which the cells in-corporate [14C]oleate into cholesteryl ['14C]oleate (21). WhereasLDL did not cause such an effect in macrophages, acetyl-LDLcaused a 100-fold increase in the incorporation of [14C]oleateinto cholesteryl ['14C]oleate (Table 2). This stimulation was in-hibited by maleyl-albumin and by fucoidin, indicating that itrequired uptake for acetyl-LDL through the high-affinitybindingz site. In the same experiment, acetyl-LDL did notstimulate the incorporation of [14C]oleate into "4C-labeled tri-glycerides or phospholipids.

3

336 Cell Biology: Goldstein et al.

Table 1. Ability of various cell types to degrade 1251-acetyl-LDL and 125I-LDL

High-affinity degradation,ng.5 hr-1 per mg protein Ratio:

Cell type Time in culturel 1251-Acetyl-LDL 1251-LDL 125I-Acetyl-LDL/1251-LDL

Mouse peritoneal macrophages 0 8,780 ± 1,310* 360 + 120* 24Rat peritoneal macrophages 1 day 33,700 190 177Guinea pig Kupffer cells 1 day 4,700 590 8.0Human monocytes 5 days 1,910 500 3.8Human lymphocytes 5 days 30 83 0.36Human fibroblasts 16 generations 15 780 0.02Human lymphoblasts >100 generations 13 422 0.03Mouse L cells >100 generations 320 1170 0.27Mouse Y-1 adrenal cells >100 generations 100 960 0.10Chinese hamster ovary cells >100 generations 0 2000 0

Cells were grown in the appropriate medium containing 10% whole serum. For experiments, cells were incubated with 1 or 2 ml of mediumcontaining 10% human lipoprotein-deficient serum and either (i) 1251-acetyl-LDL (25 Atg/ml; 50-100 cpm/ng) in the absence and presence of350 gg of unlabeled acetyl-LDL per ml or (ii) 125I-LDL (25 ,ug/ml; 50-100 cpm/ng) in the absence and presence of 350 ,g of unlabeled LDL perml. After 5 hr at 370C, the amount of 125I-labeled acid-soluble material in the medium was determined in triplicate dishes. High-affinity degradationwas calculated by subtracting the values obtained in the presence of unlabeled lipoprotein from those obtained in its absence.* Mean ± SEM for four experiments carried out on four different days.

The uptake of acetyl-LDL produced a massive increase inthe total cholesterol content of the cells after incubation for 4days (Table 3). This increase was inhibited by maleyl-albumin.Eighty percent of the cholesterol in the macrophages incubatedwith acetyl-LDL was esterified. In studies to be reported else-where, we observed that macrophages incubated.with acetyl-LDL developed large numbers of inclusions that stained withOil Red 0 and showed birefringence in polarized light, twocharacteristics of cholesteryl ester-laden macrophages invivo.The rate of uptake of 125I-acetyl-LDL in macrophages was

not increased by incubating the cells for 48 hr in lipoprotein-deficient serum, nor was it suppressed when the cells were in-cubated for 48 hr in the presence of 25-hydroxycholesterol andcholesterol, two conditions that, respectively, stimulate andsuppress the activity of the LDL receptor in fibroblasts andlymphocytes (1, 2). Moreover, prior incubation of the macro-phages with unlabeled acetyl-LDL (25 ,ug/ml) for 72 hr underconditions that caused massive cellular cholesterol deposition(Table 3) did not diminish the rate at which the macrophagestook up and degraded 125I-acetyl-LDL, indicating that the

Table 2. Stimulation of cholesteryl ester formation in mouseperitoneal macrophages incubated with acetyl-LDL

Incorporation of [14C]oleateinto cholesteryl [14C]oleate,

Addition to medium nmol/mg protein

None 0.62LDL, 25 ,ug/ml 0.83LDL, 250 ,g/ml 0.92Acetyl-LDL, 25 ,g/ml 70.0Acetyl-LDL, 25,ug/ml+ maleyl-albumin, 250,g/ml 10.6

Acetyl-LDL, 25 gg/ml+ fucoidin, 50 jg/ml 2.2

Cell monolayers were prepared by the standard procedure exceptthat 2 ml of medium A containing 5 X 106 peritoneal cells was dis-pensed into 60 X 15-mm plastic petri dishes. After the adherence step(2 hr, 37°C) and the subsequent washes, each dish received 1.5 ml ofmedium containing 10% fetal calf serum, 0.1 mM [14Cloleate-albumin(9600 cpm/nmol), and the indicated addition. After incubation at37°C for 24 hr, the monolayers were washed by the standard proce-dure (13), the cells were scraped from the dish in 1 ml of phosphate-buffered saline, and the cellular content of cholesteryl ['4C]oleate wasdetermined. Each value is the mean of duplicate incubations.

acetyl-LDL binding site was not regulated by the cellularcholesterol content.

DISCUSSIONIn the current study, we have used 125I-acetyl-LDL as an initialprobe to search for a binding site on macrophages that mediatesthe uptake and degradation of chemically altered or denaturedLDL. The studies disclosed that such a binding site was presenton mouse peritoneal macrophages as well as on macrophagesfrom several other tissues and species but not on various otherisolated cell types.The important property of the acetyl-LDL binding site is that

it facilitates the rapid uptake of the bound lipoprotein and itsdelivery to lysosomes. When macrophages bound '25I-acetyl-LDL at 4°C and were then warmed to 370C, lipoprotein deg-radation products in the form of mono[125I]iodotyrosine weredetectable in the culture medium within 10 min, and all of thelipoprotein that entered the cell was degraded within 60 min.These rapid uptake kinetics, which resemble those of the LDLreceptor in fibroblasts (1), suggest that the 125I-acetyl-LDLbinding site is functionally adapted to mediate the uptake anddegradation of molecules that bind to it. This rapid degradationaffects only the molecule that is attached to the binding site and

Table 3. Increase in cellular cholesterol content in mouseperitoneal macrophages incubated with acetyl-LDL

Total cellularcholesterol content,

Addition to medium gg sterol/mg protein*

None 28 (20%)LDL, 25 ,ug/ml 33LDL, 250 ,g/ml 55Acetyl-LDL, 25,ig/ml 1050 (80%)Acetyl-LDL, 25-tg/ml+ maleyl-albumin, 250 jtg/ml 83

Cell monolayers were prepared as described in the legend to Table2. Each dish received 2 ml ofmedium containing 10% fetal calf serumand the indicated addition (day 0). The cells were incubated for 4 daysat 37°C. The medium was replaced with fresh medium of identical-composition on days 1 and 3. On day 4 the monolayers were washedby the standard procedure (13), the cells were scraped into 1 ml ofphosphate-buffered saline, and the cellular content of cholesterol wasdetermined. Each value is the mean of triplicate incubations.* The number in parentheses is the proportion of the total cholesterolthat was esterified.

Proc. Natl. Acad. Sci. USA 76 (1979)

Proc. Natl. Acad. Sci. USA 76 (1979) 337

is not due to a ligand-mediated stimulation of nonspecific en-docytosis. This follows from the observation that unlabeledacetyl-LDL did not accelerate the uptake or degradation of125I-LDL or of 125I-labeled human serum albumin (data notshown).The binding site for 125I-acetyl-LDL in macrophages showed

high affinity and saturability. In addition to acetyl-LDL, thisbinding site recognized maleyl-LDL, maleyl-albumin, and atleast two negatively charged sulfated polysaccharides, fucoidinand dextran sulfate. These findings suggest that the binding siterecognizes, at least in part, regions of multiple negative charge.However, not all negatively charged molecules were bound tothis site. Acetyl-albumin, polyphosphates, RNA, phosvitin, andheparin failed to inhibit the binding, uptake, and degradationof 125I-acetyl-LDL.

Recent studies have shown that denatured proteins, such asalbumin treated with formaldehyde (8, 23) or nitroguanidine(9), are rapidly cleared from the circulation when injected intomice or rats. Large amounts of these proteins appear in liverKupffer cells (8, 9). Moreover, in vitro studies of rat peritonealmacrophages (24), rat yolk sac (23), and rat Kupffer cells (25)have shown rapid uptake and degradation of 125I-labeledformaldehyde-treated albumin. These studies have led to thepostulation that macrophages contain specific binding sites thatmediate the uptake of denatured proteins. The present findingswith 125I-acetyl-LDL lend support to this concept. Moreover,the formal demonstration of binding activity at 40C, compe-tition with related substances, and destruction of the bindingsite with proteolytic enzymes all suggest that, in the case of al-tered LDL, a discrete receptor molecule may be involved.The uptake of acetyl-LDL produced a large accumulation

of free and esterified cholesterol within the macrophages. Werband Cohn (26) showed that mouse peritoneal macrophagespossess high levels of lysosomal cholesterol esterase activity andthat, when macrophages ingested particulate cholesterylester-albumin complexes, the cholesteryl esters were hydro-lyzed and the free cholesterol was excreted by the cells.Whether such lysosomal hydrolysis and excretion of acetyl-LDL-derived cholesteryl esters occurs is an important questionthat remains to be answered.The physiologic significance, if any, of the acetyl-LDL up-

take system in macrophages in vivo is not yet known. However,the demonstration that uptake of acetyl-LDL through thisbinding site leads to massive cellular deposition of cholesterolraises the possibility that this receptor may be responsible forthe accumulation of LDL-derived cholesteryl esters in mac-rophages and other scavenger cells that occurs throughout thebody in patients with familial hypercholesterolemia. Supportfor this hypothesis will require the demonstration that nativeLDL can be converted in the body into a form that is recognizedby the acetyl-LDL binding site. Although in tivo acetylationof plasma LDL seems unlikely at this point, some chemical orphysical alteration of LDL occurring in plasma or interstitialfluid may make it susceptible to recognition by the macrophagebinding site.

Note Added in Proof. In experiments to be reported elsewhere, wehave observed that 125-Ilabeled acetyl-LDL injected intravenously intomice is cleared by the liver within 2 min. This in vivo hepatic uptakeprocess is blocked by fucoidin and dextran sulfate and presumablyoccurs in Kupffer cells.

Wendy Womack provided excellent assistance with the cell culturestudies. We thank Gloria Y. Brunschede and Lynda Letzig for theirable technical assistance. This work was supported by Grant PO-1-HD-20948 from the National Institutes of Health.

1. Goldstein, J. L. & Brown, M. S. (1977) Annu. Rev. Biochem. 46,897-930.

2. Goldstein, J. L. & Brown, M. S. (1978) Johns Hopkins Med. J.143,8-16.

3. Simons, L. A., Reichl, D., Myant, N. B. & Mancini, M. (1975)Atherosclerosis 21, 283-298.

4. Bilheimer, D. W., Goldstein, J. L., Grundy, S. M. & Brown, M.S. (1975) J. Clin. Invest. 56, 1420-1430.

5. Goldstein, J. L. & Brown, M. S. (1977) Metabolism 26, 1257-1275.

6. Chomette, G., De Gennes, J. L., Delcourt, A., Hammou, J. C. &Perie, G. (1971) Ann. Anat. Pathol 16,233-250.

7. Watanabe, T., Tanaka, K. & Yanai, N. (1968) Acta Pathol. Jpn.18,319-331.

8. Mego, J. L. & McQueen, J. D. (1965) Biochim. Biophys. Acta 100,136-143.

9. Buys, C. H. C. M., De Jong, A. S. H., Bouma, J. M. W. & Gruber,M. (1975) Biochim. Biophys. Acta 392,95-100.

10. Basu, S. K., Goldstein, J. L., Anderson, R. G. W. & Brown, M. S.(1976) Proc. Natl. Acad. Sci. USA 73,3178-3182.

11. Cohn, Z. A. & Benson, B. (1965) J. Exp. Med. 121, 153-169.12. Edelson, P. J. & Cohn, Z. A. (1976) in In Vitro Methods in Cell-

Mediated and Tumor Immunity, eds. Bloom, B. R. & David, J.R. (Academic Press, New York), pp. 333-340.

13. Goldstein, J. L., Basu, S. K., Brunschede, G. Y. & Brown, M. S.(1976) Cell 7,85-95.

14. Garvey, J. S. (1961) Nature (London) 191, 972-974.15. Boyum, A. (1968) Scand. J. Clin. Lab. Invest. Suppl. 21 97,

77-89.16. Johnson, W. D., Mei, B. & Cohn, Z. A. (1977) J. Exp. Med. 146,

1613-1626.17. Fraenkel-Conrat, H. (1975) Methods Enzymol. 4, 247-269.18. Glazer, A. N., Delange, R. J. & Sigman, D. S. (1975) Chemical

Modifications of Proteins (North-Holland, Amsterdam), p.79.Amsterdam), p. 79.

19. Lowry, 0. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J.(1951) J. Biol. Chem. 193, 265-275.

20. Goldstein, J. L. & Brown, M. S. (1974) J. Biol. Chem. 249,5153-5162.

21. Goldstein, J. L., Dana, S. E. & Brown, M. S. (1974) Proc. Natl.Acad. Sci. USA 71, 4288-4292.

22. Brown, M. S., Faust, J. R. & Goldstein, J. L. (1975) J. Clin. Invest.55,783-793.

23. Moore, A. T., Williams, K. E. & Lloyd, J. B. (1977) Biochem. J.164,607-616.

24. Pratten, M. K., Williams, K. E. & Lloyd, J. B. (1977) Biochem.J. 168,365-372.

25. Nilsson, M. & Berg, T. (1977) Biochim. Biophys. Acta 497,171-182.

26. Werb, Z. & Cohn, Z. A. (1972) J. Exp. Med. 135,21-44.

Cell Biology: Goldstein et al.